12

I have a vague recollection back in the late 70s of using a wonderful program called Locksmith (from memory, so I could be wrong).

It apparently could handle all the different ways that disk-based programs prevented themselves from being copied.

My memory is dim but I remember the term "nibble copier" if that's of any assistance.

I was wondering what the various copy protection methods were, how they worked, and how Locksmith (or any other method) bypassed the protections.

6
  • 8
    This subject is so broad one could write a book about it.
    – fadden
    Mar 6 at 15:32
  • 1
    @fadden I don't really feel like turning off my uMatrix protections and there's no redirect parameter that can be extracted from that rads.stackoverflow.com URL without loading it. Would you mind applying the "please no bare links" policy to your comment too and giving the title and author in plain text?
    – ssokolow
    Mar 7 at 11:53
  • 1
    @ssokolow It's just a link to a book on Amazon called "Tome Of Copy Protection" by Bruce Jones and Lane Roathe. Here's the URL in text form if that helps: www.amazon.com/Copy-Protection-Bruce-Jones/dp/1387907271 (SO automatically replaces actual Amazon links with the RADS redirect) Mar 7 at 13:25
  • 1
    @RBarryYoung Thanks. I'm in Canada, so an Amazon.ca link would be what I need, but a quick domain swap left me with a working link... that revealed it's a Lulu.com print-on-demand book, meaning better to buy it from Lulu's own storefront where a much higher amount will go to the author without Amazon taking a cut. (Though I'm less of a fan of them than I used to be. My mother received an e-mail from them about how they're starting to "retire" catalogue entries that don't get purchased for more than X amount of time.)
    – ssokolow
    Mar 7 at 14:30
  • 3
    Locksmith, Copy II Plus, Essential Data Duplicator and Disk Muncher! Those were the days!
    – Geo...
    Mar 8 at 0:30

3 Answers 3

20

That question is kind of overly broad, as it essentially asks to explain all ways a floppy image can be composed and written - which is next to infinit. A short search may turn up quite a lot of hits.

So I'd like to focus on that 'nibble copy' part:

'Bit nibbler' (*1) is a term coined in times of GCR for what today may be called 'raw copy'. It describes access to disk data based on magnetic flux changes, in sequence as they are written. While used on many platforms, it was especially popular on the Apple II, as it's disk system was almost entirely software defined. A fierce race between protection schemes and copy programs lasted over the whole time the Apple II as mainstream machine.

Lock Smith is still one of the best known ones. It was the first (widely known) program that did not copy logical data by using the standard RWTS (*2) routine, the lowest level of DOS access, but reading and reproducing the raw, GCR data.

Lock Smith made its name by constantly improving - from simply nibble wise copy all the way to track timing and orientation. But it wasn't the only one. There was a whole armada, from rather plain COPYA and COPY II all the way to NIBBLER and BACK IT UP. In fact, when copying disks - at least in a more than 'one a month' fashion - one had to use a wide variety of programs, as each had its advantages - including plain copy programs, as not every disk of a set was protected to the max. Plain (DOS based) programs had the advantage of needing less memory to hold the image, so a copy could be done faster than with programs that had to handle nibbles.

In addition, while these programs were good at automatically copying many default schemes, one still had to figure out the details for each disk, and often different schemes for different tracks.

So for ... well, let's say "less casual" operation, one had to make notes, create a data base on how to copy which disk. This was mine:

enter image description here


*1 - initially often written as 'nybbler'

*2 - Read or Write Track and Sector - the very basic routine DOS provided to access and en/decode a 256 byte sector.

2
  • I wonder if any copy protection schemes exploited the fact that the Disk II controller can read data at a fairly wide range of speeds. If one were to write a sync pattern followed by a long sequence of $AA bytes, some of which were written at a speed of 3.625us per bit (29 cycles per byte), and some of which were written at a speed of 4.375us per bit (35 cycles/byte), it would be possible to encode a bit per byte that would be totally invisible to most disk-copying software. I wonder if anyone did such a thing?
    – supercat
    Mar 6 at 19:59
  • 4
    yes, Sirius in particular made use of that fact, timing the nibbles to arrive at exact (and various) times, which could not be reproduced by a bit-copier. A bit-copier would normalise the delays to a constant time, and then the protection would fail. Datasoft also slowed their drives to allow packing more data per track than usual, requiring a fast read routine to restore it. Mar 7 at 0:47
10

I remember Locksmith, too, and I actually still have it on floppy somewhere.

There was a wide variety of protection schemes, from "let's use non-standard code for GCR address and data marks" or "let's calculate checksums in a different way" to the infamous "let's make a disk with spiral tracks, like a record".

If you want war stories, a modern-day cracker using the nick 4am has cracked quite a few, and has done a few write-ups of more interesting ones (though I am not able to google a systematic collection of the write-ups, maybe someone else can locate them).

I guess if anyone would be able to give a good summary of protection methods used, it would be him.

Basically Locksmith and other nibbler copy programs worked by reading the raw data of a complete track, and then writing it without converting it to actual data, circumventing any protection scheme that relied on non-standard encoding.

Additionally, they could display the raw it so you could figure out oddities and make manual adjustments.

2
  • 3
    Every 4am crack has an accompanying text file that explains to process. Some background into the what and how: Confessions of a Disk Cracker: the secrets of 4am. and a2-4am / passport, 4am's own disk deprotector. I'm still amazed that using blank sectors to give different results on every read worked on the Apple II.
    – scruss
    Mar 6 at 15:22
  • When linking to archive.org never link directly to iaXXXXX.archive.org servers. those links might break. Take the link that you get from the File list that starts with archive.org/download/. that one will link you to the current iaXXXX server since files might get shuffled around.
    – masterX244
    Mar 7 at 12:39
8

You can find some detailed descriptions below. There are two primary classes of protections that Locksmith could not duplicate - "E7" and "weak bits", the first of which was used to protect Locksmith itself!

A Brief Description of Some Popular Copy-Protection Techniques on the Apple ][ Platform - PoC||GTFO 0x10 (part 1 part 2)

and

In Search of the Most Amazing Thing; or, Towards a Universal Method to Defeat E7 Protection on the Apple II Platform (pdf) - PoC||GTFO 0x11 (3)

1
  • 2
    Locksmith used copy protection? Well, that's just rude! :-)
    – paxdiablo
    Mar 7 at 1:34

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.